It is well established that cardiac contractility is impaired in patients with type I and II diabetes mellitus as well as in experimental animals with genetic or drug-induced diabetes. The mechanism(s), however, for these functional and possibly metabolic defects are unclear. The working hypothesis of this proposal is that exercise training of diabetic animals will improve cardiac contractile performance and metabolism. By determining the metabolic mechanisms through which exercise training improves the function of the diabetic hearts new information about the cause of diabetes-related cardiac dysfunction will be obtained. These studies represent the initial steps toward the long-term goal of preventing the decline in myocardial function that often occurs in diabetic patients, perhaps through moderate exercise training. The effects of chronic diabetes mellitus and exercise training on myocardial contractile performance, energy metabolism and substrate utilization will be assessed in two models of diabetes: the streptozotocin-induced diabetic rat and the genetically obese Zucker rats. The streptozotocin-induced diabetic rats have many of the characteristics of type I diabetes, while the obese Zucker rats are similar to the type II form of diabetes. In all studies, four groups of rats will be investigated: sedentary control, trained control, sedentary diabetic and trained diabetic. For the studies on the streptozotocin-induced diabetic rat a dose of 55 mg/kg, i.v. of streptozotocin will be used and the duration of diabetes will be eight weeks. Treatment of lean and genetically obese Zucker rats will be initiated at 2 months of age and will be continued for 6 months. The effects of diabetes and exercise training on cardiac contractile activity will be studied by measuring in vivo basal myocardial left intraventricular pressure and dP/dt, the responsiveness to i.v. norepinephrine, and by constructing Frank-Starling curves from isolated perfused working hearts. The vulnerability of the isolated perfused heart to global ischemia will also be assessed. Hearts will be perfused with concentrations of glucose and free fatty acids which characterize the in vivo state. Dynamic changes in high energy phosphate metabolites will be measured in Langendorff perfused hearts as a function of aortic perfusion pressure using 31 P-nuclear magnetic resonance spectroscopy. Fatty acid and glucose utilization and tissue content of long-chain acyl carnitine and Co-A esters will also be measured. The metabolic parameters will be correlated to the cardiac contractile activity. Although the metabolic factors that contribute to the functional disturbances that develop in the diabetic heart have not been elucidated unequivocally, abnormal substrate utilization and energy metabolism are implicated as probable sites of action. By reversing of reducing the severity of the metabolic alterations, exercise training may prevent the development of diabetic cardiomyopathy.